Wasserstein Dictionary Learning: Optimal Transport-based unsupervised non-linear dictionary learning

Morgan A. Schmitz (), Matthieu Heitz (), Nicolas Bonneel (), Fred Ngolè () and David Coeurjolly ()
Additional contact information
Morgan A. Schmitz: Astrophysics Department; IRFU; CEA; Université Paris-Saclay
Matthieu Heitz: Université de Lyon; CNRS; LIRIS
Nicolas Bonneel: Université de Lyon; CNRS; LIRIS
Fred Ngolè: LIST, Data Analysis Tools Laboratory, CEA Saclay
David Coeurjolly: Université de Lyon; CNRS; LIRIS

No 2017-84, Working Papers from Center for Research in Economics and Statistics

Abstract: This article introduces a new non-linear dictionary learning method for histograms in the probability simplex. The method leverages optimal transport theory, in the sense that our aim is to reconstruct histograms using so called displacement interpolations (a.k.a. Wasserstein barycenters) between dictionary atoms; such atoms are themselves synthetic histograms in the probability simplex. Our method simultaneously estimates such atoms, and, for each datapoint, the vector of weights that an optimally reconstruct it as an optimal transport barycenter of such atoms. Our method is computationally tractable thanks to the addition of an entropic regularization to the usual optimal transportation problem, leading to an approximation scheme that is e cient, parallel and simple to differentiate. Both atoms and weights are learned using a gradient-based descent method. Gradients are obtained by automatic di erentiation of the generalized Sinkhorn iterations that yield barycenters with entropic smoothing. Because of its formulation relying on Wasserstein barycenters instead of the usual matrix product between dictionary and codes, our method allows for non-linear relationships between atoms and the reconstruction of input data. We illustrate its application in several different image processing settings. ;Classification-JEL: 33F05, 49M99, 65D99, 90C08

Keywords: Optimal Transport; Wasserstein barycenter; Dictionary Learning (search for similar items in EconPapers)
Pages: 40 pages
Date: 2017-08-07
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
http://crest.science/RePEc/wpstorage/2017-84.pdf CREST working paper version (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:crs:wpaper:2017-84

Access Statistics for this paper

More papers in Working Papers from Center for Research in Economics and Statistics Contact information at EDIRC.
Bibliographic data for series maintained by Secretariat General () and Murielle Jules Maintainer-Email : murielle.jules@ensae.Fr.